SOPHIA ANTIPOLIS, France– Median Technologies (Paris:ALMDT) (ALMDT:PA) announces outstanding performance based on a large-scale lung cancer screening (LCS) patient cohort to characterize malignant vs benign lung nodules based on a deep learning algorithm on low-dose chest computed tomography. Median plans to demonstrate how the potential of its iBiopsy AI powered Computer Aided Diagnosis (CADx) product for LCS could significantly impact the accuracy, consistency, and adoption of LCS worldwide. The Median Technologies iBiopsy® Lung Cancer Development Plan was launched in February 2021, as part of the Company’s strategic positioning in early-stage disease diagnosis.
Lung cancer is the number one cancer killer globally, with an estimate of 1.8m deaths, accounting for 25% of all cancer deaths in 2018. This is largely due to late-stage diagnosis of the disease associated with poor 5-year survival. Major international studies have shown that LDCT lung cancer screening can reduce mortality by 44% (NELSON Trial, 2018). Despite being included in US screening guidelines and the increased adoption by the medical community, screening efforts still suffer from low rates of adoption. The low patient enrollment is commonly attributed to the difficulty to characterize malignant from benign nodules leading to high rates of false positives, unnecessary follow up procedures and raising concerns on the economic viability of screening programs.
The study was based on a cohort of 1,696 patients from the National Lung Screening Trial cases (NLST) consisting of a total of 15,608 lung nodules. The training set was based on 1,224 patients with 11,392 nodules and the test set was based on 472 patients having a total of 4,216 nodules.
The performance of iBiopsy® CADx for the characterization of lung nodules shows an AUC of 0.991 and an outstanding sensitivity of 95.2% for a specificity of 95.7% and is a major progress for Lung Cancer Screening worldwide.
“We have achieved an outstanding performance with our innovative deep learning algorithm for lung nodule characterization. This promising and powerful digital biomarker could have a profound impact on lung cancer screening and help to dramatically reduce the 1.8m lung cancer deaths per year by diagnosing lung cancer patients at a stage that could save their lives.”, highlights Fredrik Brag, CEO and founder of Median. “Our plan in the coming months is to perform an additional large-scale study for an end-to-end lung cancer screening digital biomarker. Results of this study including both nodule detection (CADe) and characterization (CADx) are expected in Q4, 2021”, Brag added.
Median’s AI digital biomarker could significantly contribute to eliminate barriers to widespread LCS adoption. It could have a huge impact on saving patients’ lives for the deadliest form of cancers as well as providing substantial savings by eliminating unnecessary healthcare spendings.
